نوع مقاله : مقاله مستخرج از رساله دکتری

نویسندگان

1 دانشجوی دکتری رشته حسابداری، گروه حسابداری ، دانشکده اقتصاد و حسابداری، دانشگاه آزاد اسلامی واحد تهران جنوب، تهران، ایران.

2 استادیارعلوم اقتصادی، گروه اقتصاد نظری، دانشکده اقتصاد و حسابداری، دانشگاه آزاد اسلامی واحد تهران مرکز، تهران، ایران

3 استادیار حسابداری، گروه حسابداری، دانشکده اقتصاد و حسابداری، دانشگاه آزاد اسلامی واحد تهران جنوب، تهران، ایران.

4 دانشیار حسابداری، گروه حسابداری، دانشکده اقتصاد و حسابداری، دانشگاه آزاد اسلامی واحد تهران جنوب، تهران، ایران.

چکیده

هدف این پژوهش تحلیل مقایسه‌ای دقت پیش بینی روش های محاسباتی ریسک بازار در ارزش در معرض خطر با رویکرد هوش مصنوعی ارتباطی می‌باشد توسعه روز افزون بازارهای مالی اهمیت برآورد معیار شناخته شده اندازه گیری ریسک بازار، (ارزش در معرض خطر) را بیش از گذشته آشکار ساخته است. ارزش در معرض خطر (Var) یک معیار آماری است که حداکثر زیان مورد انتظار از نگهداری یک دارایی یا پرتفوی را در دوره زمانی معین و با احتمال مشخص (سطح اطمینان معلوم) محاسبه و به صورت کمی گزارش می‌کندو یکی از مهم‌ترین معیارهای ریسک بازار است که به‌طور گسترده برای مدیریت ریسک مالی توسط نهادهای قانون‌گذار مالی و مدیران پرتفوی به‌کاربرده می‌شود. ریسک‌ها در سطح کلان دارای آثار فراگیر هستند و می‌توانند تأثیرات منفی را در کل بازار مالی برجای بگذارند. شناخت وابستگی های درونی و ارتباطات متقابل شرکت ها و توسعه معیارهای ریسک که افزایش وابستگی دنباله بازده شرکت ها را در طول بحران را پیش بینی نماید ، از اهمیت زیادی برخوردار است. وجود چنین روش هایی، یک ابزار قدرتمند به منظور افزایش ثبات مالی آتی در اختیار تصمیم گیران قرار می‌دهد . بدین‌جهت با استفاده از اطلاعات روزانه قیمت سهام ، ارزش در معرض خطر با روش‌های پارامتریک (روش واریانس –کوواریانس)، شبیه‌سازی تاریخی، شبیه‌سازی بوت استرپ بین دوره زمانی 1390 الی 1396 بورس اوراق بهادار تهران برای شرکت‌های نمونه آماری ، محاسبه و استفاده‌شد. پس از کاهش نوسانات روش Bootstrap، Historical و Variance covariance با استفاده از تبدیل موجک برای آموزش مدل‌ها و پیش بینی، روش هر 15 روز متوالی را به‎عنوان ورودی (همان متغیر مستقل) در مدل RVM و روز 16 ام به‎عنوان متغیر وابسته را به‎عنوان خروجی مدل در نظر گرفته شدو برای ارزیابی مدل‌ها از دو معیار ارزیابی بانام‌های میانگین مربعات خطا (MSE)، میانگین قدر مطلق خطا (MAE) استفاده‌شده است برای پیش‌بینی از الگوریتم ماشین بردار ارتباطی استفاده‌شده است. الگوریتم RVM یک مدل غیرخطی است و با انتقال داده ها از فضای ورودی به فضای ویژگی باعث غیرخطی شدن الگوریتم می‌شود. در ماشین بردار ارتباطی از کرنل گوسی برای غیرخطی سازی استفاده‌شده است. نتایج آزمون فرضیه‌ها و برازش الگوریتم هوش مصنوعی ارتباطی نشان داد که الگوریتم هوش مصنوعی جهت پیش‌بینی روش‌های روزانه ارزش در معرض خطر روش کارایی می باشد و همچنین در بازار سرمایه ایران پیش‌بینی ارزش در معرض خطر با روش نیمه پارامتریک بوت استرپ باقدرت بالاتری انجام و جهت استفاده توصیه می گردد، روش های پارامتریک(واریانس - کوواریانس) و شبیه سازی تاریخی در رتبه‌های بعدی قرار می‌گیرند. مطالعات انجام‌شده در مورد ارزش در معرض ریسک محدود به یک صنعت و یا با تعریف پرتفویی بوده است و تمام شرکت‌های بورسی موردبررسی قرار نگرفته‌اند، در این مطالعه سعی شد تمام شرکت‌های حاضر در بورس ریسک بازارشان با رویکرد ارزش در معرض ریسک تحت 3 مدل مهم و پرکاربرد واریانس –کواریانس، شبیه‌سازی تاریخی، شبیه‌سازی بوت استرپ محاسبه شود و با استفاده از الگوریتم هوش مصنوعی کارایی آن‌ها سنجیده شود. به‌نوعی پژوهش‌های پیشین از جامعه آماری کمتر و عدم سنجش کارایی مدل‌ها در عمل برخوردارند

کلیدواژه‌ها

موضوعات

عنوان مقاله [English]

PREDICTING VALUE AT RISK: AN ARTIFICIAL INTELLIGENCE APPROACH

نویسندگان [English]

  • mohammad zamani 1
  • Ghodratollah Emamverdi 2
  • Yadollah Noorifard 3
  • Mohsen Hamidian 4
  • Seyedeh Mahboubeh Jafari 3

1 PhD Candidate in Accounting, Department of Accounting, Faculty of Economics and Accounting, Islamic Azad University, South Tehran Branch, Tehran, Iran.

2 Assistant Professor of Economics, Department of Theoretical Economics, Faculty of Economics and Accounting, Islamic Azad University, Tehran Central Branch, Tehran, Iran

3 Assistant Professor of Accounting, Department of Accounting, Faculty of Economics and Accounting, Islamic Azad University, South Tehran Branch, Tehran, Iran.

4 Associate Professor of Accounting, Department of Accounting, Faculty of Economics and Accounting, Islamic Azad University, South Tehran Branch, Tehran, Iran.

چکیده [English]

The purpose of this study is to compare the accuracy of predicting market risk calculation methods of value at risk with the relevance of the artificial intelligence approach. the increasing development of financial markets has revealed the importance of estimating the well-known measure of market risk, risk value more than before. Value at Risk (VaR) is a statistical measure that calculates and quantifies the maximum expected loss from holding an asset or portfolio over a period of time with a certain probability (known confidence level) and is one of the most important market risk criteria that is widely used to manage financial risk by financial regulators and portfolio managers. Macro-level risks have pervasive effects and can have negative effects on the entire financial market. Recognition of the interdependence and mutual relevance of companies and the development of risk factors that predict an increase of dependence on the sequence returns to the company during the crisis is of utmost importance. The existence of such methods provides a powerful tool for decision-makers to increase future financial stability. After reducing the fluctuations of Bootstrap, Historical, and Variance covariance with using wavelet transformation for model training and forecasting, the method is used every 15 consecutive days as input (same as the independent variable) in RVM model and 16th day as the dependent variable was considered as the model output and to evaluate the models, two evaluation criteria called Mean Square Error (MSE) and Mean Absolute Error (MAE) have been used. The relevance vector machine algorithm was used to predict variables. The RVM algorithm is a nonlinear model and causes the algorithm to be nonlinear by transferring data from the input space to the property space. The reason for checking with the MAE error is that this error represents the mean of the absolute value and is more comprehensible for us than the MSE, which is the Mean Squared Error. The results of testing the hypotheses and fitting the relevant artificial intelligence algorithm showed that the artificial intelligence algorithm is an efficient method for predicting daily value-at-risk methods. Also, in the Iranian capital market, risk-value forecasting is done with the semi-parametric bootstrap method with higher power and is recommended for use. Parametric methods (variance-covariance) and historical simulation are in the next ranks. Studies on value at risk have been limited to one industry or by portfolio definition and all listed companies have not been examined.
In this study, we tried to calculate the market risk of all companies listed on the stock exchange with the value-at-risk approach under 3 important and widely used models: variance-covariance, historical simulation, bootstrap simulation and measure their efficiency using an artificial intelligence algorithm. In a way, previous researches have a less statistical population and do not measure the efficiency of models in practice.

کلیدواژه‌ها [English]

  • Market risk
  • value at risk
  • Communication Artificial Intelligence Algorithm
Abdelghany, K. E. (2005). Disclosure of market risk or accounting measures of risk: an empirical study. Managerial Auditing Journal, 25, 867-875.
Alexander, C. (2009). Market risk analysis, value at risk models (Vol. 4). John Wiley & Sons.
Angelidis, T., & Degiannakis, S. (2005). Modeling risk for long and short trading positions. The Journal of Risk Finance, 6(3), 226-238.
Assaf, A. (2015). Value-at-Risk analysis in the MENA equity markets: Fat tails and conditional asymmetries in return distributions. Journal of Multinational Financial Management, 29, 30-45.
Atefi, E., & Ranjbar, M. R. (2019). Estimation Value at Risk using by combining approach Exteme Value Theory and CIPRA at Tehran stock Exchange. Financial Engineering and Portfolio Management, 38(10), 375-394. dor:20.1001.1.22519165.1398.10.38.17.3 [in persian]
Barone-Adesi, G., Giannopoulos, K., & Vosper, L. (1999). VaR without correlations for nonlinear portfolios. Journal of Futures Markets, 19, 583-602.
Barone‐Adesi, G., & Giannopoulos, K. (2000). Non parametric Value-at-Risk techniques. myths and realities. Economic Notes, 30(2), 167-181.
Bauwens, L., Hafner, C. M., & Laurent, S. (2012). Handbook of volatility models and their applications (Vol. 3). John Wiley & Sons.
Biek Khormizi, M., & Rafei, M. (2020). Modeling Value at Risk of Futures Contract of Bahar Azadi Gold Coin with Considering the Historical Memory in Observations Application of FIAPARCH-CHUNG Models. Journal of Asset Management and Financing8(1), 57-82. doi: 10.22108/amf.2018.107307.1189 (in Persian)
Bijelic, A., & Ouijjane, T. (2019). Predicting Exchange Rate Value-at-Risk and Expected Shortfall: A Neural Network Approach.
Botshekan, M., Peymani, M., & Sadredin Karami, M. (2019). Estimate and evaluate non-parametric value at risk and expected shortfall based on principal component analysis in Tehran Stock Exchange. Financial Management Perspective, 8(24), 79-102. dor: 20.1001.1.26454637.1397.8.24.4.2 (in Persian)
Butler, J., & Schachter, B. (1997). Estimating value-at-risk with a precision measure by combining kernel estimation with historical simulation. Review of Derivatives Research, 1, 371-390.
Darabi, R., Vaghfi, S. H., & Salmanian, M. (2017). Relationship between social responsibility reporting with company value and risk for companies registered in Tehran Stock Exchange. Valued and Behavioral Accountings Achievements, 1(2), 193-213.doi:  ‎ 10.18869/acadpub.aapc.1.2.193 (in persian)
Ebrahimi, S. B., Aghaei, M., & Mohebbi, N. (2017). Estimating Portfolio Value-at-Risk and Expected Shortfall by Possibility and Necessity Theory. Financial Research Journal, 19(2), 193-216.doi: 10.22059/jfr.2017.218621.1006298 (in persian)
Echaust, K., & Just, M. (2020). Value at risk estimation using the GARCH-EVT approach with optimal tail selection. Mathematics, 8(1), 114.
Eqbalnia, M. (2008). Testing the value at risk model for forecasting and managing investment risk. Business Management Perspectives, 21, 33-54.
Fallahshams, M., Naserpour, A., Saqafi, A., & Taqavifard, M. T. (2017). The Use of Incremental Value at Risk (IVaR) in Calculating Portfolio Risk Using “Before and After. Strategic Management Thought, 11(2), 205-226. doi:  10.30497/smt.2017.2159 (in persian)
Fallahshams, M. F., Naserpour, A., Saqafi, A., & Taqavifard, M. T. (2017). The Use of Incremental Value at Risk (IVaR) in Calculating Portfolio Risk Using “ Before and After" Approach. Strategic Management Thought11(2), 205-226. doi: 10.30497/smt.2017.2159 (in persian)
Fereydoni, Farshid, Darabi, Roya, Anvar Rostami, Ali Asghar. (2020). Application of artificial intelligence algorithm in predicting profit smoothing. Financial Accounting and Auditing Research, 12 (45), 103-134. https://civilica.com/doc/1045483(in persian)
Ghaffari, F., Nikomram, H., & Zomordian, G. (2014). Study of the ability to explain neural network models in measuring the value at risk. Journal of Financial Engineering and portfolio Management, 5(19), 19-38. dor: 20.1001.1.22519165.1393.5.19.2.5 (in persian)
Ghulam, Y., & Doering, J. (2017). Spillover effects among financial institutions within Germany and the United Kingdom. Research in International Business and Finance, 44, 49-63.
Hamidian, M., Habibzadeh Baygi, S. J., Salmanian, M., & Vaghfi, S. H. (2016). The Systematic Risk Prediction of Listed Companies in Tehran Stock Exchange Using Ant Colony and LARS Algorithm. Journal of Iranian Accounting Review, 3(10), 19-40. doi: 0.22055/jiar.2016.12732 (in persian)
He, K., Ji, L., Tso, G. K., Zhu, B., & Zou, Y. (2018). Forecasting exchange rate value at risk using deep belief network ensemble based approach. Procedia computer science, 139, 25-32.
Heidari Haratmeh, M. (2019). Portfolio Optimization with CVaR under VG Process. Financial Knowledge of Securities Analysis, 12(41), 101-112. magiran.com/p1959212 (in persian)
Joaquin, D. C. (2016). On animal spirits and economic decisions: Value-at-Risk and Value-within-Reach as measures of risk and return. The Quarterly Review of Economics and Finance, 60, 231-233.
Jorion, P. (2000). Value at Risk: The New Benchmark for Managing Financial Risk. European financial management, 6(3), 277-300.
Kachecha, C., & Strydom, B. (2011). Using Accounting Data as a Measure of Systematic Risk.
Mohammad Zadeh, A., & Masoud Zadegan, S. (2017). Forecasting Daily Volatility and Value at Risk with High Frequency Data. Journal of Development & Evolution Mnagement, 1395(27), 63-74. available at: https://civilica.com/doc/792026(in persian)
Nabavi Chashmi, S. A., Ghanbari Memeshi, E., & Memarian, E. (2018). Value at Risk in Tehran Stock Exchange using Non-parametric and parametric Approaches. Business Management, 46, 252-272.available at: magiran.com/p2149817 (in persian)
Naderi Nooreini, M. M. (2018). The Best Methodology of Estimation of Value-at-Risk in Iranian Mutual Funds. Journal of Asset Management and Financing, 6(1), 159-180.doi:  10.22108/amf.2017.21353 (in persian)
Narimani, R., Hakimipour, N., & Rezaei, A. (2013). Application of artificial neural network method and conditional heterogeneity variance models in calculating the risk value. Financial Economics, 7(24), 101-137.dor: 20.1001.1.25383833.1392.7.24.4.9 (in persian)
Patton, A. J., Ziegel, J. F., & Chen, R. (2019). Dynamic semiparametric models for expected shortfall (and value-at-risk). Journal of econometrics, 211(2), 388-413.
Paytakhti Oskooe, S. A., Hadipour, H., & Aghamiry, H. (2019). The Stock Optimal Portfolio using value at risk: Evidence from Tehran Stock Exchange. Empirical Studies in Financial Accounting, 15(61), 157-178.
Pritsker, M. (2006). The hidden dangers of historical simulation. Journal of Banking & Finance, 30(2), 561-582.
Raghfar, H., & Ajorlo, N. (2016). Calculation of Value at Risk of Currency Portfolio for a Typical Bank by GARCH-EVT-Copula Method. Iranian Journal of Economic Research, 21(67), 113-141. doi: https://doi.org/10.22054/ijer.2016.7238
Rahnamarodposhti, F., Ghandehari, S., & Sharareh. (2015). Estimating of value at risk - based risk assessment on the performance evaluation of active portfolio management in tehran stock exchange. Financial engineering and portfolio management, 6(24), 91.dor: 20.1001.1.22519165.1394.6.24.6.6 (in persian)
Rastgoo, N., & panahian, h. (2018). Designing and Explaining the Systematic Risk Estimation Model using metaheuristic Method in Tehran Stock Exchange: Adaptive Approach to the Model of Econometrics and Artificial Intelligence. Financial Engineering and Portfolio Management, 35(9), 19-49. doi:20.1001.1.22519165.1398.10.41.11.3 [in persian]
Rezagholizadeh, M., elmi, Z., & mohammadi majd, S. (2023). The Effect of Financial Stress on the Stock Return of Accepted Industries in Tehran Stock Exchange. Quarterly Journal of Quantitative Economics (JQE)20(1), 32-73. doi: 10.22055/jqe.2021.35405.2284
Sajjad, R., & Taherifar, R. (2016). Confidence interval Calculation & Evaluating Markov regime switching Precision for Value-at-Risk Estimation: A Case Study on Tehran Stock Exchange Index (TEDPIX). Financial Research Journal, 18(3), 461-482. doi: 10.22059/jfr.2016.62451 (in persian)
Salehi, M., Mousavi Shiri, M., & Ebrahimi Swizi, M. (2014). The information content of declared dividends per share and predicted earnings per share in explaining abnormal stock return. 21(6), 117-140. dor: 20.1001.1.23830379.1393.6.21.5.5 (in persian)
Sener, F., Bas, C., & Ikizler-Cinbis, N. (2012). On recognizing actions in still images via multiple features. European Conference on Computer Vision,
Shafiee, A., Abdoh, T. H., Raei, R., & Falahpor, S. (2019). Estimation of Value at Risk with Extreme Value Theory approach and using Stochastic Differential Equation. 10(40), 325-348.dor: 20.1001.1.22519165.1398.10.40.15.5 (in persian)
Talibnia, G., & Ahmadi Nezamabadi, F. (2010). Investigating the Predictive Power of the Famav French (F&F) Three-Factor Model and the Value at Risk (VaR) Model in Selecting the Optimal Stock Portfolio of Companies Listed on the Tehran Stock Exchange. Journal of Management Accounting, 3(6), 49-62. available at: https://sanad.iau.ir/Journal/jma/Article/816531 (in persian)
Taylor, J. W. (2020). Forecast combinations for value at risk and expected shortfall. International Journal of Forecasting, 36(2), 428-441.
Tehrani, R., Mohammadi, S., & Porebrahimi, M. (2011). Modeling and forecasting the volatility of Tehran Exchange Dividend Price Index (TEDPIX). Financial Research Journal, 12(30), 23-36 dor: 20.1001.1.10248153.1389.12.30.1.1 (in persian)
Tipping, M. E. (2000). The relevance vector machine. Advances in neural information processing systems, Exchange. Quarterly Journal of Quantitative Economics (JQE), 19(4), 43-78.
Torki, L., Esmaeli, N., & Haghparast, M. (2023). Comparison of GARCH Family Models in Estimating Value at Risk and Conditional Value at Risk on the Tehran Stock Exchange. Quarterly Journal of Quantitative Economics (JQE)19(4), 43-78. doi: 10.22055/jqe.2021.33186.2240 (in persian)
Zhang, D., Sikveland, M., & Hermansen, Ø. (2018). Fishing fleet capacity and profitability. Marine Policy, 88, 116-121. Doi: https://doi.org/10.1016/j.marpol.2017.11.017